Dupree and colleagues (1995) conducted a retrospectively assembled case-control study of lung cancer among employees in four uranium processing operations. Two of the groups were employed at the uranium processing plant at the Oak Ridge facility during nonoverlapping time periods and have been previously described in the studies by Polednak and Frome (1981) and by Checkoway and colleagues (1988). The third operation was the Mallinckrodt Chemical Works Uranium Division in Missouri, and the fourth was the Feed Materials Production Center in Fernald, Ohio (described in the study by Ritz, 1999). The primary exposure in common among these facilities was to alpha radiation from airborne dust containing insoluble natural uranium compounds. Enriched uranium was present at the Oak Ridge facility from 1943 to 1946, and exposure to radium and radon daughters was also possible at the Mallinckrodt facility. All operations, except the Oak Ridge facility from 1943 to 1947, also processed thorium.
The authors identified eligible cases of lung cancer (n = 787) by mortality follow-up of the employee cohorts through the end of 1982, which provided at least 30 years of follow-up for each cohort. One control was selected for each case, matching on race, gender, and birth and hire dates within 3 years. The authors required both cases and controls to have been employed for at least 6 months. By focusing on a limited number of cases and controls rather than the entire cohort, they were able to re-create each person’s work history in detail to provide a quantitative individual estimate of exposure to radiation.
Using employment and occupational radiation monitoring records, data was collected, as available, on smoking history (limited to never or ever smoked because further details were not routinely available), socioeconomic status (as first pay code), and complete work history. Smoking data were obtained for 48 percent of the cases and 39 percent of the controls, with 91 percent of the cases and 75 percent of the controls identified as smokers. Using the available individual and environmental data (including air monitoring data), health physicists estimated annual radiation lung doses from deposited uranium for each person.
As discussed throughout this section there is overlap between some of the cohorts of uranium-processing workers. Table 4.4 points out the extent of overlap.
This section discusses the scientific literature on the potential associations between human health effects and exposure to uranium. The committee has reorganized the results of the population studies (described in the previous section) and other research into disease-specific subsections that discuss the results and the strengths and limitations of current knowledge. The focus of the discussion is on the organs and organ systems that are the principal sites of uranium deposition following exposure. Malignant and nonmalignant diseases of the lung are impor-